Dual-Use AI Risk

Dual-use AI risk refers to the possibility that an AI system built for legitimate purposes can also be turned toward harmful ones, such as generating convincing phishing campaigns, automating vulnerability discovery for attackers, or lowering the skill barrier for synthetic fraud. For enterprises, this matters because a capable model a company licenses or deploys can create liability it did not intend, from misuse by employees or contractors to misuse by outside parties who gain access to its outputs. Governance teams need to assess these misuse pathways in vendor due diligence, acceptable-use policies, and internal access controls, since regulators increasingly expect organizations to anticipate foreseeable misuse of the tools they put into service.

1 item

ResearchGlobal2026-11-03

ARTEX Multi-Agent Framework Used to Hit 200 Orgs in State-Linked Campaign

CrowdStrike researchers documented a China-nexus group called Vapor Panda using ARTEX, an open-source multi-agent AI framework, to automate attacks. More than 200 organizations in 14 countries were targeted starting October 17, 2026. Government, telecom, pharma, and defense organizations were the primary targets.

DeepSeek agentic AI dual-use AI risk cyber espionage open-weight models China AI incident CrowdStrike